- GreatSQL社区原创内容未经受权不得随便应用，转载请分割小编并注明起源。
[toc]
明天介绍MGR集群的日常治理保护操作，包含主节点切换，单主&多主模式切换等。手工操作以及利用MySQL Shell两种形式都会别离介绍。
当初有个三节点的MGR集群：
mysql> select * from performance_schema.replication_group_members;+---------------------------+--------------------------------------+--------------+-------------+--------------+-------------+----------------+| CHANNEL_NAME | MEMBER_ID | MEMBER_HOST | MEMBER_PORT | MEMBER_STATE | MEMBER_ROLE | MEMBER_VERSION |+---------------------------+--------------------------------------+--------------+-------------+--------------+-------------+----------------+| group_replication_applier | af39db70-6850-11ec-94c9-00155d064000 | 172.16.16.10 | 3306 | ONLINE | PRIMARY | 8.0.25 || group_replication_applier | b05c0838-6850-11ec-a06b-00155d064000 | 172.16.16.11 | 3306 | ONLINE | SECONDARY | 8.0.25 || group_replication_applier | b0f86046-6850-11ec-92fe-00155d064000 | 172.16.16.12 | 3306 | ONLINE | SECONDARY | 8.0.25 |+---------------------------+--------------------------------------+--------------+-------------+--------------+-------------+----------------+1. 切换主节点
当主节点须要进行保护时，或者执行滚动降级时，就能够对其进行切换，将主节点切换到其余节点。
在命令行模式下，能够应用 group_replication_set_as_primary() 这个udf实现切换，例如：
-- 将Primary角色切换到第二个节点mysql> select group_replication_set_as_primary('b05c0838-6850-11ec-a06b-00155d064000');+--------------------------------------------------------------------------+| group_replication_set_as_primary('b05c0838-6850-11ec-a06b-00155d064000') |+--------------------------------------------------------------------------+| Primary server switched to: b05c0838-6850-11ec-a06b-00155d064000 |+--------------------------------------------------------------------------+1 row in set (1.00 sec)[root@yejr.run:mysql.sock] [(none)]>select * from performance_schema.replication_group_members;+---------------------------+--------------------------------------+--------------+-------------+--------------+-------------+----------------+| CHANNEL_NAME | MEMBER_ID | MEMBER_HOST | MEMBER_PORT | MEMBER_STATE | MEMBER_ROLE | MEMBER_VERSION |+---------------------------+--------------------------------------+--------------+-------------+--------------+-------------+----------------+| group_replication_applier | af39db70-6850-11ec-94c9-00155d064000 | 172.16.16.10 | 3306 | ONLINE | SECONDARY | 8.0.25 || group_replication_applier | b05c0838-6850-11ec-a06b-00155d064000 | 172.16.16.11 | 3306 | ONLINE | PRIMARY | 8.0.25 || group_replication_applier | b0f86046-6850-11ec-92fe-00155d064000 | 172.16.16.12 | 3306 | ONLINE | SECONDARY | 8.0.25 |+---------------------------+--------------------------------------+--------------+-------------+--------------+-------------+----------------+顺便提一下，在MySQL 5.7版本中，只能通过重启以实现主节点的主动切换，不能手动切换。从这个角度来说，如果想要应用MGR，最好是抉择MySQL 8.0版本，而不要应用5.7版本。
如果是用MySQL Shell，则能够调用 setPrimaryInstance() 函数进行切换：
#首先获取cluster对象 MySQL 172.16.16.10:3306 ssl JS > var c=dba.getCluster()#查看以后各节点列表 MySQL 172.16.16.10:3306 ssl JS > c.status() "status": "OK", "statusText": "Cluster is ONLINE and can tolerate up to ONE failure.", "topology": { "172.16.16.10:3306": { "address": "172.16.16.10:3306", "memberRole": "PRIMARY", "mode": "R/W", "readReplicas": {}, "replicationLag": null, "role": "HA", "status": "ONLINE", "version": "8.0.25" }, "172.16.16.11:3306": { "address": "172.16.16.11:3306", "memberRole": "SECONDARY", "mode": "R/O", "readReplicas": {}, "replicationLag": null, "role": "HA", "status": "ONLINE", "version": "8.0.25" }, "172.16.16.12:3306": { "address": "172.16.16.12:3306", "memberRole": "SECONDARY", "mode": "R/O", "readReplicas": {}, "replicationLag": null, "role": "HA", "status": "ONLINE", "version": "8.0.25" } }, "topologyMode": "Single-Primary" }, "groupInformationSourceMember": "172.16.16.10:3306"}#执行切换 MySQL 172.16.16.10:3306 ssl JS > c.setPrimaryInstance('172.16.16.11:3306')Setting instance '172.16.16.11:3306' as the primary instance of cluster 'MGR1'...#列举了三个节点各自产生的变动Instance '172.16.16.10:3306' was switched from PRIMARY to SECONDARY.Instance '172.16.16.11:3306' was switched from SECONDARY to PRIMARY.Instance '172.16.16.12:3306' remains SECONDARY.WARNING: The cluster internal session is not the primary member anymore. For cluster management operations please obtain a fresh cluster handle using dba.getCluster().#实现切换The instance '172.16.16.11:3306' was successfully elected as primary.2. 切换单主/多主模式
在命令行模式下，能够调用 group_replication_switch_to_single_primary_mode() 和 group_replication_switch_to_multi_primary_mode() 来切换单主/多主模式。
#间接调用函数即可mysql> select group_replication_switch_to_multi_primary_mode();+--------------------------------------------------+| group_replication_switch_to_multi_primary_mode() |+--------------------------------------------------+| Mode switched to multi-primary successfully. |+--------------------------------------------------+#查看各节点状态mysql> select * from performance_schema.replication_group_members;+---------------------------+--------------------------------------+--------------+-------------+--------------+-------------+----------------+| CHANNEL_NAME | MEMBER_ID | MEMBER_HOST | MEMBER_PORT | MEMBER_STATE | MEMBER_ROLE | MEMBER_VERSION |+---------------------------+--------------------------------------+--------------+-------------+--------------+-------------+----------------+| group_replication_applier | af39db70-6850-11ec-94c9-00155d064000 | 172.16.16.10 | 3306 | ONLINE | PRIMARY | 8.0.25 || group_replication_applier | b05c0838-6850-11ec-a06b-00155d064000 | 172.16.16.11 | 3306 | ONLINE | PRIMARY | 8.0.25 || group_replication_applier | b0f86046-6850-11ec-92fe-00155d064000 | 172.16.16.12 | 3306 | ONLINE | PRIMARY | 8.0.25 |+---------------------------+--------------------------------------+--------------+-------------+--------------+-------------+----------------+#切换成单主模式时能够指定某个节点的 server_uuid，如果不指定则会依据规定主动抉择一个新的主节点#在这里，我抉择了指定mgr3节点作为新主mysql> select group_replication_switch_to_single_primary_mode('b0f86046-6850-11ec-92fe-00155d064000');+-----------------------------------------------------------------------------------------+| group_replication_switch_to_single_primary_mode('b0f86046-6850-11ec-92fe-00155d064000') |+-----------------------------------------------------------------------------------------+| Mode switched to single-primary successfully. |+-----------------------------------------------------------------------------------------+在MySQL Shell中，能够调用 switchToSinglePrimaryMode() 以及 switchToMultiPrimaryMode() 函数进行切换。同样地，函数 switchToSinglePrimaryMode() 里也能够指定某个节点作为新的主节点。
MySQL 172.16.16.10:3306 ssl JS > var c=dba.getCluster()#切换到多主模式 MySQL 172.16.16.10:3306 ssl JS > c.switchToMultiPrimaryMode() #切换到单主模式，这里我指定mgr2节点作为新主 MySQL 172.16.16.10:3306 ssl JS > c.switchToSinglePrimaryMode("172.16.16.11:3306")留神，在曾经是单主模式时，无论是 group_replication_switch_to_single_primary_mode() 还是 switchToSinglePrimaryMode() 函数中指定另一个节点时是不会产生切换的，但也不会报错，只有提醒。
3. 增加新节点
接下来咱们演示如何向MGR集群中增加一个新节点。
首先，要先实现MySQL Server初始化，创立好MGR专用账户、设置好MGR服务通道等前置工作，这部分的操作能够参考前文 3. 装置部署MGR集群。
接下来，间接执行命令 start group_replication 启动MGR服务即可，新增的节点会进入分布式复原这个步骤，它会从已有节点中主动抉择一个作为捐献者（donor），并自行决定是间接读取binlog进行复原，还是利用Clone进行全量复原。
如果是曾经在线运行一段时间的MGR集群，有肯定存量数据，这时候新节点退出可能会比较慢，倡议手动利用Clone进行一次全量复制。还记得后面创立MGR专用账户时，给加上了 BACKUP_ADMIN 受权吗，这时候就排上用场了，Clone须要用到这个权限。
上面演示如何利用Clone进行一次全量数据恢复，假设要新增的节点是 172.16.16.13 （给它命名为 mgr4）。
#在mgr4上设置捐献者#为了升高对Primary节点的影响，倡议抉择其余Secondary节点mysql> set global clone_valid_donor_list='172.16.16.11:3306';#停掉mgr服务（如果有的话），敞开super_read_only模式，而后开始复制数据#留神这里要填写的端口是3306（MySQL失常服务端口），而不是33061这个MGR服务专用端口mysql> stop group_replication; set global super_read_only=0; clone INSTANCE FROM GreatSQL@172.16.16.11:3306 IDENTIFIED BY 'GreatSQL';全量复制完数据后，该节点会进行一次主动重启。重启结束后，再次确认 group_replication_group_name、group_replication_local_address、group_replication_group_seeds 这些选项值是否正确，如果没问题，执行 start group_replication 后，该节点应该就能够失常退出集群了。
如果是用MySQL Shell增加新节点则更简略。先执行MySQL Server初始化，并执行 dba.dba.configureInstance() 创立MGR专用账号后。而后，连贯到Primary节点，间接调用 addInstance() 函数即可：
#连贯到Primary节点$ mysqlsh --uri GreatSQL@172.16.16.10:3306 MySQL 172.16.16.10:3306 ssl JS > var c=dba.getCluster() MySQL 172.16.16.10:3306 ssl JS > c.addInstance('GreatSQL@172.16.16.13:3306')WARNING: A GTID set check of the MySQL instance at '172.16.16.13:3306' determined that it contains transactions that do not originate from the cluster, which must be discarded before it can join the cluster. ...NOTE: 172.16.16.13:3306 is being cloned from 172.16.16.10:3306 #<--主动抉择一个donor节点** Stage DROP DATA: Completed** Clone Transfer FILE COPY ############################################################ 100% Completed PAGE COPY ############################################################ 100% Completed REDO COPY ############################################################ 100% CompletedNOTE: 172.16.16.13:3306 is shutting down...* Waiting for server restart... ready* 172.16.16.13:3306 has restarted, waiting for clone to finish...** Stage RESTART: Completed* Clone process has finished: 72.43 MB transferred in about 1 second (~72.43 MB/s)Incremental state recovery is now in progress.* Waiting for distributed recovery to finish...NOTE: '172.16.16.13:3306' is being recovered from '172.16.16.12:3306'* Distributed recovery has finished#新节点胜利退出结束The instance '172.16.16.13:3306' was successfully added to the cluster.#确认增加胜利，已在MGR集群列表中 MySQL 172.16.16.10:3306 ssl JS > c.describe(){ "clusterName": "MGR1", "defaultReplicaSet": { "name": "default", "topology": [ { "address": "172.16.16.10:3306", "label": "172.16.16.10:3306", "role": "HA" }, { "address": "172.16.16.11:3306", "label": "172.16.16.11:3306", "role": "HA" }, { "address": "172.16.16.12:3306", "label": "172.16.16.12:3306", "role": "HA" }, { "address": "172.16.16.13:3306", <--新退出的节点 "label": "172.16.16.13:3306", "role": "HA" } ], "topologyMode": "Single-Primary" }} 确认新节点增加胜利。
4. 删除节点
在命令行模式下，一个节点想退出MGR集群，间接执行 stop group_replication 即可，如果这个节点只是长期退出集群，前面还想加回集群，则执行 start group_replication 即可主动再退出。而如果是想彻底退出集群，则进行MGR服务后，执行 reset master; reset slave all; 重置所有复制（蕴含MGR）相干的信息就能够了。
在MySQL Shell里，只需调用 removeInstance() 函数即可删除某个节点，例如：
MySQL 172.16.16.10:3306 ssl JS > c.removeInstance('172.16.16.13:3306');The instance will be removed from the InnoDB cluster. Depending on the instancebeing the Seed or not, the Metadata session might become invalid. If so, pleasestart a new session to the Metadata Storage R/W instance.Instance '172.16.16.13:3306' is attempting to leave the cluster...The instance '172.16.16.13:3306' was successfully removed from the cluster.这就将该节点踢出集群了，并且会重置 group_replication_group_seeds 和 group_replication_local_address 两个选项值。之后该节点如果想再退出集群，须要调用 addInstance() 从新加回。
5. 异样退出的节点从新加回
当节点因为网络断开、实例crash等异常情况与MGR集群断开连接后，这个节点的状态会变成 UNREACHABLE，待到超过 group_replication_member_expel_timeout + 5 秒后，集群会踢掉该节点。等到这个节点再次启动并执行 start group_replication，失常状况下，该节点应能主动从新加回集群。
在MySQL Shell里，能够调用 rejoinInstance() 函数将异样的节点从新加回集群：
MySQL 172.16.16.10:3306 ssl JS > c.rejoinInstance('172.16.16.13:3306'); Rejoining instance '172.16.16.13:3306' to cluster 'MGR1'...The instance '172.16.16.13:3306' was successfully rejoined to the cluster.6. 重启MGR集群
失常状况下，MGR集群中的Primary节点退出时，剩下的节点会主动选出新的Primary节点。当最初一个节点也退出时，相当于整个MGR集群都敞开了。这时候任何一个节点启动MGR服务后，都不会主动成为Primary节点，须要在启动MGR服务前，先设置 group_replication_bootstrap_group=ON，使其成为疏导节点，再启动MGR服务，它才会成为Primary节点，后续启动的其余节点也能力失常退出集群。可自行测试，这里不再做演示。
P.S，第一个节点启动结束后，记得重置选项 group_replication_bootstrap_group=OFF，防止在后续的操作中导致MGR集群决裂。
如果是用MySQL Shell重启MGR集群，调用 rebootClusterFromCompleteOutage() 函数即可，它会主动判断各节点的状态，抉择其中一个作为Primary节点，而后拉起各节点上的MGR服务，实现MGR集群重启。能够参考这篇文章：万答#12，MGR整个集群挂掉后，如何能力主动选主，不必手动干涉
7. 小结
本文介绍了MGR集群几种常见治理保护操作方法，包含切换主节点，切换单主/多主模式，增加节点，删除节点，异样节点重退出，重启整个MGR集群等。总的来看，利用MySQL Shell治理MGR集群会更简略不便些，也有利于治理平台的封装，不过手工操作的形式也不能遗记，有些时候可能没有配套的MySQL Shell工具，就得靠手工了。
参考资料、文档
- MySQL 8.0 Reference Manual
- 数据库内核开发 - 温正湖
- Group Replication原理 - 宋利兵
免责申明
因集体程度无限，专栏中不免存在错漏之处，请勿间接复制文档中的命令、办法间接利用于线上生产环境。请读者们务必先充沛了解并在测试环境验证通过前方可正式施行，防止造成生产环境的毁坏或侵害。
Enjoy GreatSQL :)
本文由博客一文多发平台 OpenWrite 公布！